Experimental evidence for acceleration of reaction between iodine monoxide and chlorine monoxide at the reactor surface

Article abstract of DOI:10.1023/A:1023300428051

The reaction of iodine monoxide with chlorine monoxide resulting in atom escape to the gas phase is studied at T = (303 ± 5) K and P = 2.5 Torr using a flow setup for measuring the resonance fluorescence signals of atomic iodine and chlorine. The heterogeneous reaction between chlorine monoxide and iodine monoxide occurring at the reactor surface covered with an F32-L Teflon-like compound and treated by the reaction products is characterized by the rate constant k = (4.9 ± 0.2) x 10^-11 cm³ molecule^-1 s^-1. This value is substantially higher than the rate constant for the homogeneous reaction IO^. + ClO^. (k₁ ≤ 1 x 10^-12 cm³ molecule^-1 s^-1).

Full text of DOI:10.1023/A:1023300428051

Kinetics and Catalysis, Vol. 44, No. 2, 2003, pp. 202–210. Translated from Kinetika i Kataliz, Vol. 44, No. 2, 2003, pp. 218–226.
Original Russian Text Copyright © 2003 by Larin, Spasskii, Trofimova, Turkin.
Experimental Evidence for Acceleration of Reaction
between Iodine Monoxide and Chlorine Monoxide
at the Reactor Surface
I. K. Larin, A. I. Spasskii, E. M. Trofimova, and L. E. Turkin
Institute of Energy Problems of Chemical Physics, Russian Academy of Sciences, Moscow, 117829 Russia
Received June 10, 2002
Abstract—The reaction of iodine monoxide with chlorine monoxide resulting in atom escape to the gas phase
is studied at T = (303 ± 5) K and P = 2.5 Torr using a flow setup for measuring the resonance fluorescence sig-
nals of atomic iodine and chlorine. The heterogeneous reaction between chlorine monoxide and iodine monox-
ide occurring at the reactor surface covered with an F32-L Teflon-like compound and treated by the reaction prod-
–
.
11
3
–1 –1
ucts is characterized by the rate constant k = (4.9 ± 0.2) × 10 cm molecule s . This value is substantially
.
–
12
3
–1 –1
higher than the rate constant for the homogeneous reaction IO + ClO (k ≤ 1 × 10 cm molecule s ).
1
of 10–10 cm molecule s at the above-mentioned con-
3
–1 –1
INTRODUCTION
centration of active iodine.
In 1977, Bejanian et al. [7] measured the rate con-
stant for the reaction of IO with ClO at 298 K by
The possible role of reactions of active iodine
.
(
iodine atoms and IO radicals) in the processes result-
.
.
ing in the ozone layer depletion has been actively dis-
cussed during the last decade. Earlier, iodine com-
pounds were believed not to reach the stratosphere, that
is, a height of 20–30 km, where most of the ozone is
mass spectrometry using a discharge flow system:
–
11
3
–1 –1
k = (1.1 ± 0.2) × 10 cm molecule s .
Three thermochemically acceptable channels of this
located, because their lifetime in the atmosphere is reaction were considered:
measured in several days or weeks and because they
rapidly decompose in chemical and photochemical
reactions at a low height (0–10 km). However, more
recently [1, 2] it has been shown that convective fluxes
can transport water-insoluble compounds very rapidly
from the lower height to the upper troposphere or even
to the stratosphere. Gaseous radon, which is formed
exclusively near the earth’s surface and has as a short
lifetime as 3.8 days, was detected over the tropical
tropopause [3].
.
.
.
IO + ClO
I + OClO ,
(
Ia)
∆
H = –(4.0 ± 0.2) kcal/mol,
.
.
IO + ClO
I + Cl + O2,
(
Ib)
∆
H = –(2.3 ± 1.2) kcal/mol,
.
.
IO + ClO
H = –(47.5 ± 1.2) kcal/mol.
The fractions of pathways (Ia), (Ib), and (Ic) were
ICl + O ,
2
(
Ic)
∆
It is known [4] that biomass is rich in iodine. Biom-
ass is mainly burned in the tropical regions with strong
convective fluxes, which transport iodine-containing
compounds to the stratosphere. McCormick [5]
observed a dramatic depletion of the ozone layer in the
(
0.55 ± 0.03), (0.25 ± 0.02), and (0.2 ± 0.02), respec-
tively.
.
.
The rate constant for the reaction of IO with ClO
stratosphere at a height of less than 20 km not only in was also measured in [8] using an induced resonance
polar regions but also at middle latitudes and even in fluorescence laser in a flow system at 200–362 K. The
.
tropical regions, where the concentrations of active bro- kinetics of IO radical consumption was monitored in
.
mine and, especially, chlorine are much lower than that
of active iodine [6]. He explained ozone depletion at
low and middle latitudes observed during the last
decade by the presence of a small concentration of
active iodine (≤1pptv), which reacts with the chlorine
and bromine compounds. If ozone depletion were
excess of the second reactant ClO , that is, under con-
ditions of a pseudomonomolecular process. The fol-
lowing expression for this rate constant was derived:
k(T) = (5.1 ± 1.7)
× 10–12exp[(280 ± 80)/T] cm molecule s .
At 298 K,
k = (1.17 ± 0.08) × 10–11 cm molecule s .
3
–1 –1
.
.
exclusively due to the reaction of IO and ClO result-
ing in atomic iodine and chlorine or their precursors,
the rate constant of this reaction would be of the order
3
–1 –1
0
023-1584/03/4402-0202$25.00 © 2003 MAIK “Nauka /Interperiodica”

EXPERIMENTAL EVIDENCE FOR ACCELERATION OF REACTION
203
The fraction of the pathways that do not produce atomic
iodine was (0.14 ± 0.04) at 298 K.
EXPERIMENTAL
Experimental Setup
The results of the above two studies agree perfectly,
To study the reaction of iodine monoxide and chlo-
whereas the measured rate constants were nearly an rine monoxide, a setup was constructed that allowed the
order of magnitude lower than that required to explain resonance fluorescence signals of atomic iodine and
the ozone loss in the lower stratosphere observed dur- chlorine to be measured simultaneously. Measurements
ing the last decade.
were conducted at (303 ± 5) K and a pressure inside the
reactor of 2.5 Torr. The setup contained a reactor,
Moreover, the IO.
concentration in the strato- sources of iodine and chlorine atoms, systems for the
sphere was lower than 1 pptv and equal to at most registration of the iodine and chlorine atoms, and a sys-
+
0.3
tem for the supply of gases into the reactor.
0
.2– pptv [9].
0.2
The flow setup, the sources of iodine and chlorine
atoms, the systems for the registration of the iodine and
chlorine atoms, and the systems for the supply of gases
into the reactor were described in detail in [14]. There-
.
The above data suggest that the reaction of IO with
.
ClO proposed in [6] cannot lead to a decrease in the
ozone concentration in the lower stratosphere, because fore, we will present here only a brief description of
reactions responsible for this process should occur them, giving more attention to reactor modifications.
much more rapidly.
Two reactor modifications were used for the studies.
The reactor of modification (a) is schematically
shown in Fig. 1. The inner surface of the reactor was
covered with an F32-L Teflon-like compound to reduce
the heterogeneous decay of atoms and radicals. Oxy-
gen, ozone, helium, and ethane were introduced into
the reactor through side holes. A source of iodine atoms
was installed inside the reactor. The distance between
the inlet of iodine atoms and their registration zone was
It is known that surface reactions are usually faster
than reactions in the gas phase. This may be due to both
an increase in the reaction rate constant (due to a
decrease in the activation energy) and an increase in the
local concentration of active species. It is the heteroge-
neous reactions in polar stratospheric clouds that cause
the formation of ozone holes in the north hemisphere in
spring. The reactions occurring on surfaces such as glass
and Teflon, as well as on ice particles, are much more
rapid than those occurring in the gas phase [10, 11]. The
introduction of these processes into the computer mod-
els of the atmosphere resulted in an increase in the cal-
culated ozone loss by an order of magnitude. Many
researchers [12] suggested that chemical reactions
could also occur on dust and soot particles liberated
8
.5 cm. It was possible to supply chlorine atoms to any
point along the reactor axis through a mobile nozzle.
The mobile nozzle consisted of a quartz tube with an
outer diameter of 6 mm and a wall thickness of 1 mm.
The inner surface of the nozzle was covered with
hydrocarbon grease to reduce the heterogenous decay
of chlorine atoms. A collapsible Teflon connection
enabled one to move the nozzle without depressurizing
the reactor.
during volcano eruption. When studying the reactions
.
of IO radicals with a number of sulfur-containing
Another reactor, modification (b), differed from the
first one in that it had a mobile source of iodine atoms
and an immobile source of chlorine atoms. This reactor
compounds, we also showed [13] that heterogeneous
reactions between these components are much faster
than the same reactions in the gas phase.
was used to study changes in the concentration of
.
When we measured the rate constant for iodine atom iodine atoms or IO radicals with changes in the con-
.
formation during the reaction of the IO radicals with tact time.
.
Two different substances were used to produce
iodine atoms in various experiments. Iodine atoms were
generated by the photolysis of a CH I (or C F I) mix-
ClO [14], we noticed that, as the reaction vessel walls
were treated with the reaction products, the reaction
rate constant increased until it attained a constant value,
thus indicating the possibility of the heterogeneous pro-
cess.
3
3 7
ture diluted with oxygen (0.1–0.5% of CH I or C F I in
3
3 7
O ). The surface of a flow chamber of photolysis was
2
treated with orthophosphoric acid to reduce the hetero-
In this work, we present the results of experiments geneous decay of iodine atoms. Photolysis was con-
carried out in the reaction vessel with the walls pre- ducted at λ = 253.7 nm.
treated with an F32-L Teflon-like compound and then
Molecular oxygen was stored in commercial metal-
lic cylinders under pressures of 150–100 Torr and sup-
plied to the reactor using a system ensuring the constant
.
.
with the products of the IO with ClO reaction.
The aim of this work was to determine whether the flow rate accurate to 2% for several hours. The starting
reaction under study is homogeneous or not under our reactants for iodine atom production ëH I and C F I
3
3 7
operating conditions and, if not, to determine the ratio (reagent grade) were stored in a dark glass cylinder in
of the rates of the homogeneous and heterogeneous the liquid phase. The ëH I or C F I vapors were intro-
3
3 7
pathways of this reaction under these conditions.
duced through a fine-control valve into an oxygen flow
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

2
04
LARIN et al.
Cl + He
typical concentration of iodine atoms in the experi-
2
Microwave
discharge
C H
2 6
10
3
ments was ~3 × 10 atom/cm .
Chlorine atoms were generated with a discharge of
frequency 254 MHz and an intensity of 2.5 W in a
Pressure gauge port
O + O
2
3
Cl flow diluted with helium.
2
The resonance-fluorescence technique with photon
counting was also used to detect chlorine atoms. The
source of the resonance emission consisted of a flow
CH I + O
3
2
1
3
Mercury
lamp
Cl
lamp with a mixture of Cl and helium (1 × 10 and
2
17
3
1
× 10 molecule/cm , respectively). The lamp was
activated by a 254-MHz discharge. A homemade pho-
toionization counter was used to detect the emission.
The counter was filled with a mixture of argon and
nitrogen monoxide and operated at λ = 133.8 nm. MgF2
windows of the lamp and the counter enabled us to use
I
Registration zone the 119-nm chlorine line for registration, that is, to
work in the spectral region virtually free of the absorp-
tion of molecular oxygen, which was present in the
reactor. The computer system for the accumulation and
Vacuum pump port
Fig. 1. Reactor of modification (a) for studying the reaction
.
.
analysis of the chlorine signal was similar to that
employed for iodine. A signal/noise ratio equal to 2 was
of IO and ClO radicals.
1
0
3
attained at [Cl] = 1 × 10 atom/cm . The typical chlo-
rine atom concentration in our experiments was
passing through the iodine atom source. The flow rate
was determined by measuring the pressure decrease in
a calibrated volume. Chlorine was stored in glass cylin-
ders and added through a capillary to a helium flow,
which then passed the flow resonance lamp acting as a
chlorine atom source. The molecular chlorine flow was
1
2
13
3
1
0 −10 atom/cm .
EXPERIMENTAL METHOD
.
.
The method of investigating the IO + ClO reac-
varied by changing the Cl pressure at the capillary
2
tion is based on the chain process involving two chain
propagation reactions:
inlet.
Ethane used to calibrate the absolute sensitivity of
the system to chlorine atoms was stored in a glass cyl-
inder and added directly to the reactor through a capil-
lary.
the reaction studied
.
.
k1
IO + ClO
and the reaction
I + O3
I + products
(I)
Ozone was produced in an O flow passing an ozo-
2
.
k2
nator. Oxygen from the glass cylinder was fed into a
flowmeter with the constant pressure equal to 1 Torr at
its outlet. Upon purification in a low-temperature trap
IO + O .
(II)
2
If only two reactions, (I) and (II), occur in the reac-
with dry ice, oxygen passed a flow-control valve and tor, the steady-state concentration of iodine atoms [I]ss
entered an Ozon-2 ozonator designed at the Institute of is achieved if the time of the reactant contact is longer
Chemical Physics of the Russian Academy of Sciences than the characteristic time of reaction (II). This steady-
(
Moscow). Then an oxygen–ozone mixture passed state concentration can be found by the equation
through a glass cell (length, 10 cm) placed inside the
spectrometer to measure the ozone concentration by the
change in the optical density at λ = 253.7 nm.
.
.
k [O ][I] = k [ClO ][IO ]ss
.
(1)
.
2
3
ss
1
.
Here, [IO ] is the steady-state concentration of IO
radicals. If there are no chain termination reactions of
ss
The atom resonance fluorescence technique with
photon counting was used to detect iodine atoms. A
.
flow lamp with a mixture of helium and molecular iodine atoms or IO radical decay, the following rela-
1
3
17
3
iodine vapors (1 × 10 and 1 × 10 molecule/cm , tion is true:
respectively) was employed as a source of resonance
emission. A resonance lamp activated by a 254-MHz
discharge emitted at λ = 178 nm. A photoionization
.
[
I] + [IO ] = [I] ,
(2)
ss
ss
0
where [I] is the initial concentration of iodine atoms in
the reactor. Combining Eqs. (1) and (2), we arrive at
0
counter was constructed in our laboratory to detect pho-
tons. The signal from the counter was sent to a com-
puter for storage and further analysis. The accumula-
tion time was 60 s. The signal/noise ratio was 2 at an
[I]
k [O ]
2 3
0
-
-------- = --------------------- + 1.
(3)
.
[
I]
ss
8
3
k [ClO ]
iodine atom concentration [I] ~1 × 10 atom/cm . The
1
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

EXPERIMENTAL EVIDENCE FOR ACCELERATION OF REACTION
205
.
.
The straight line representing the [I] /[I] vs. [O ]/[ClO ] into the reactor transformed into ClO radicals in the
0
3
dependence should intercept the y axis at the point fast reaction
equal to unity ([I] /[I] = 1). From the slope of this line,
.
0
k3
Cl + O3
ClO + O .
2
(III)
one can determine the k /k ratio and calculate the k
2
1
1
.
The rate constant for this reaction is k = 1.2 ×
value using the rate constant for the reaction of IO
3
−
11
.
3
–1 –1
with ozone (k ) obtained earlier in a wide temperature 10 cm molecule s at 295 K [16]. Therefore, the
2
range [15].
ClO concentration was equal to the concentration of
Equation (3) is only applicable to homogeneous the chlorine atoms introduced and, in our experiments,
1
2
13
3
reaction (I) if its every step results in the formation of changed from 1 × 10 to –1 × 10 molecule/cm . The
iodine atoms.
technique for measuring the chlorine atom concentra-
tion is described in [14]. The reaction of IO and ClO
.
.
radicals (reaction (I)) resulted in the formation of a
steady-state concentration of iodine atoms.
Measurement of the Initial Resonance Fluorescence
Signal of Iodine Atoms in the Reactor—J0
The flows of é and He through the iodine atom source,
2
the ozonator, and the capillary were the same as in the
experiment. The pressure in the reactor was P = 2.5 Torr,
and the linear flow rate was v = 450–500 cm/s. The con-
RESULTS AND DISCUSSION
Taking into account that the initial concentration of
iodine atoms introduced into the reactor [I] is propor-
10
3
0
centration of iodine atoms was ~3 × 10 molecule/cm
and practically did not change along the reactor axis.
The rate constant for the heterogeneous decay of
iodine atoms was measured in special experiments
tional to the measured difference between signals (J – F)
0
and that the steady-state concentration of iodine atoms
[
I] is proportional to the signal difference (J – F),
ss
Eq. (3) can be rearranged into the form
–
1
and found equal to k ≤ 1 s . The contact time was at
–
2
most 2 × 10 s; therefore, the concentration of iodine
atoms in the registration zone was equal to their initial
concentration within ±2%. The initial resonance fluo-
rescence signal of iodine atoms was 500–1000 counts
per 10 s in different experiments.
J – F k [O ]
0
2
3
-
----------- -- = --------------------- + 1.
(4)
.
J – F
k [ClO ]
1
Recall that we derived Eq. (3) for a completely homo-
geneous reaction with the generation of iodine atoms
by every step of the process. In this case, a plot of
.
Measurement of the Background Signal—F
(J – F)/(J – F) versus [O ]/[ClO ] should be a straight
0
3
line that intersects the Y axis at y = 1.
Upon measuring and recording the initial reso-
nance fluorescence signal of iodine atoms; more than
Figure 2 presents the (J – F)/(J – F) vs.
0
1
4
3
.
4
× 10 molecule/cm of ozone was introduced into the
[
O ]/[ClO ] plot obtained in the experiments carried
3
reactor without changing other parameters of the flow.
out in the reactor, which was treated with the reaction
.
This resulted in the fast formation of IO radicals via
products many times, at a constant ozone concentration
.
the reaction of iodine with ozone (II).
and a variable [ClO ] concentration. The experimental
The rate constant for reaction (II) is 1.1 ×
points are extrapolated with a straight line, which, how-
ever, intersects the y axis at a point that is well above
y = 1. This is an indication of the heterogeneous nature
12
3
–1 –1
1
0− cm molecule s at 295 K [15] and the ozone
concentrations used. The characteristic time of the
–
3
reaction was at most 2 × 10 s. Therefore, more than of the reaction of the reactants and the possible decay
.
of active chain carries in chain termination reactions.
9
9.9% of iodine atoms transformed into IO radicals
.
.
under the experimental conditions.
As noted above, the reaction of IO with ClO can
Thus, the background signal was largely associated occur via different pathways, one of which (Ic) results
with the light scattering on the molecules of substances, in chain termination. If during the contact, reaction (Ic)
which were present in the reactor (mainly on CH I).
occurs to a significant extent, the steady-state signal of
3
iodine atoms is lower than the initial one even at
.
[
O ]/[ClO ]
0 and relation (4) is not fulfilled.
3
Measurement of the Steady-State Resonance
Fluorescence Signal of Iodine Atoms—J
We considered both of the above possibilities. First,
we studied the possibility of the heterogeneous reac-
tion. Taking into account that heterogeneous processes
are faster than homogeneous ones, we hypothesized
that the rate constant (higher than that reported in [7, 8])
When the initial resonance fluorescence signal
almost disappeared, that is, when almost all the iodine
atoms introduced into the reactor transformed into IO.
radicals, chlorine atoms produced by a discharge were of the reaction studied was due to the occurrence of a
added to the reactor.As before, the reactant flow param- heterogeneous reaction along with a homogeneous one.
eters were kept constant. Chlorine atoms introduced The technique used allowed us to distinguish these two
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

2
06
LARIN et al.
(J0 – F)/(J – F)
the registration zone is not cylindrical and that the
whole reactor surface is not available for active species.
This coefficient was chosen in [13] to be 0.5 when the
theoretical curves were fitted into the experimental
data.
1
2
1
0
9
8
7
6
5
4
3
2
1
0
1
1
Note that the study of the heterogeneous reaction
mechanism was not our task. We tried to show that the
reaction, whose rate depends on the reactant concentra-
tions, as well as the rate of the bimolecular homoge-
neous reaction, still occurs at the surface. This explains
the fact that the rate constants of the same reaction mea-
sured in different studies sometimes differ by orders of
het
magnitude. Clearly, the rate constant k1 considered
here can only be used under our experimental condi-
tions and in the given reactor.
It follows from Eq. (5) that the [I] /[I] vs.
0
ss
.
[
O ]/[ClO ] plot also intersects the y axis at the point
3
.
y = 1 if experiments are performed at a rather low ClO
.
het
1
00
200
30•0
concentration when k1 [ClO ] Ӷ 0.5k . In this case,
diff
[
O ]/[ClO ]
3
the rate constant of the reaction determined from the
slope of the curve is a sum of the rate constants for the
homogeneous and heterogeneous reactions. If experi-
.
Fig. 2. (J – F)/(J – F) vs. [O ]/[ClO ] plot obtained in
experiments performed at a constant ozone concentration
equal to 4.38 × 10 molecule/cm and a variable [ClO ]
concentration. The dotted line corresponds to a 98% confi-
dence interval.
0
3
ments are conducted at a constant and relatively low
.
.
1
4
3
ClO concentration, the reaction rate constant deter-
mined on the basis of their results might not be the rate
constant of the homogeneous process. Obviously, this
.
.
explains why our rate constant for the IO + ClO
processes. In the case of the chain process in the reac- reaction is higher than reported values [14].
tor, the characteristic time of every elementary step
The experimental curve presented in Fig. 2 was
may be shorter than that of active species diffusion to
the wall, thus permitting one to distinguish the homo-
geneous and heterogeneous reaction pathways.
obtained under conditions when the ozone concentra-
.
tion remained unchanged and the ClO concentration
changed broadly. The steady-state concentration of
10
11
3
iodine atoms was 10 –10 molecule/cm .
Consideration of the Heterogeneous
.
.
We determined the sum of the rate constants for the
heterogeneous and homogeneous reactions and, then,
using mathematical modeling based on Eq. (5),
obtained the fraction of the homogeneous component
in the reaction rate constant. The simulation results
indicated the heterogeneous character of the reaction.
Reaction IO + ClO
het
If heterogeneous reaction (k1 ) with the rate con-
het
stant k1 occurs in the reactor along with homoge-
neous one (I), Eq. (3) should be rewritten in the form
We conducted experiments varying the ozone con-
k
[O ]
3
.
[
I]
0
2
-
-------- = ------------------------------------------------------------------ ---------------- + 1 , (5) centration in the course of the run with a constant ClO
.
[
I]
0.5k
diff
ss
het
concentration. The results of these experiments are
[
ClO ]
k + k -----------------------------------------------
1
1
.
het
1
given in Fig. 3. It presents the (J – F)/(J – F) vs.
0
k [ClO ] + 0.5k
diff
.
.
[
O ]/[ClO ] plot obtained at two different ClO con-
3
.
het
1
2
where kdiff = 23D/4R , k [ClO ] is the rate constant centrations, as well as the results of mathematical mod-
eling using the rate constant of the heterogeneous reac-
tion calculated from the data of Fig. 2.
of the kinetically controlled reaction, D is the diffusion
.
coefficient of IO radicals, and R is the reactor diame-
.
1
2
ter (cm). D = D (760/P), where D is the diffusion coef-
0
0
Curve 1a corresponds to [ClO ] = 4.5 × 10 mole-
.
het
ficient of IO radicals at 760 Torr and P is the reactor
pressure (Torr). D = 0.12 cm /s at T = 296 K.
3
= 4.9 × 10–11 cm molecule s ,
3
–1 –1
cule/cm and k
1
2
.
0
12
whereas curve 2a corresponds to [ClO ] = 1.4 × 10 mol-
ecule/cm and k = 4.9 × 10–11 cm molecule s .
The fact that the coefficient at kdiff in Eq. (5) is lower
than unity suggests that the reactor configuration near
het
1
3
3
–1 –1
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

EXPERIMENTAL EVIDENCE FOR ACCELERATION OF REACTION
207
As can be seen, neither of these curves adequately
(J – F)/(J – F)
0
describes the experimental data. It is also clear that this
dependence cannot be extrapolated by a straight line
that intercepts the y axis at the point y = 1. Note that the
10
9
8
7
6
5
4
3
2
1
0
1
a
1
b
2
a
2
b
curves passing the extremum points intercept the y axis
.
at y < 1 at [O ]/[ClO ]
0. This suggests that the
3
steady-state signal of iodine atoms is stronger than the
initial signal; that is, the system contains an additional
iodine atom source.
In our opinion, chlorine atoms produced by reac-
tion (1b) in our reactor can enter either the reaction
with ozone (III) or the reaction with methyl iodide:
k4
Cl + CH3I
I + products.
(IV)
1
2
At low ozone concentrations, reaction (IV) prevails
over reaction (III) and an additional iodine atom source
appears in the reactor.
We included reaction (IV) into the chain process
scheme and performed computer simulation for the
experimental data given in Fig. 3. In this case, curve 1b
1
00
200
300
•
[
O ]/[ClO ]
3
.
corresponds to [ClO ] = 4.5 × 10 molecule/cm3,
12
.
het
–11
3
–1 –1
Fig. 3. (J – F)/(J – F) vs. [O ]/[ClO ] plot obtained at vari-
0 3
k1 = 4.9 × 10 cm molecule s , and k = 4.0 ×
4
0−12 cm molecule s , whereas curve 2b corresponds
3
–1 –1
able [O ] and constant [ClO ] concentration equal to
3
1
(1) 4.5 × 1012 and (2) 1.4 × 1012 molecule/cm . The points
correspond to the experiment, and the lines correspond to
the results of the mathematical modeling with due regard
for (1a, 2a) the heterogeneous pathway of reaction (I) and
3
.
het
1
–12
12
3
to [ClO ] = 1.4 × 10 molecule/cm , k
= 4.9 ×
3
0−11 cm molecule s , and k = 4.0 × 10 cm mol-
3
–1 –1
1
4
–
1 –1
ecule s .
As can be seen, the results of computer simulation
perfectly agree with the experimental data at these val-
(1b, 2b) the simultaneous occurrence of the heterogeneous pro-
het
cess and reaction (IV); k1 = 4.9 × 10–11 cm molecule s ,
3
–1 –1
.
k4 = 4.0 × 10–12 cm molecule s , (1a, 1b) [ClO ] =
3
–1 –1
het
ues of k1 and k . The experimental estimate of the rate
4.5 × 10 , and (2a, 2b) 1.4 × 10 molecule/cm3.
12
12
4
constant of reaction (IV) obtained more recently was
k4 = (3.1 ± 1.2) × 10–12 cm molecule s .
3
–1 –1
was used as an iodine atom source. Figure 5 shows that
our experimental data can be attributed to both the het-
erogeneous character of reaction (I) and chain termina-
tion (Ic) with the rate constant
To eliminate the effect of reaction (IV), we used
7
C F I instead of methyl iodide as a source of iodine
3
atoms. In preliminary experiments, we showed that no
iodine atoms were formed during the reaction of C F I
3
7
with chlorine atoms. Figure 4 presents the experimental
–11
3
–1 –1
k = 1.5 × 10 cm molecule s .
1
c
1
4
data obtained with C F I at [é ] = 5.0 × 10 mole-
3
7
3
3
cule/cm , P = 2.54 Torr, v = 488 cm/s, and T = 303 K,
as well as the results of mathematical modeling includ-
To distinguish between these two cases, we studied
how the resonance fluorescence signal of atomic iodine
changed with the time of reactant contact. If the rate
het
ing the heterogeneous process (I ).
Figure 4 shows that our experimental data largely constant for reaction (Ic) is low, the reaction
correspond to the heterogeneous reaction pathway.
.
However, the same data may be interpreted assuming
that chain termination (Ib) occurs at a noticeable rate
along with reaction (Ic).
k5
IO + wall
(IO)ads
(V)
.
is slow and the active iodine concentration ([I] + [IO ])
does not noticeably change during the contact time, the
This should result in a decrease in the active iodine
.
concentration ([I] + [IO ]) along the reactor. The exper- resonance fluorescence signal of iodine atoms should
imental iodine concentration [I] should be lower than be independent of the contact time for both the homo-
that predicted by Eq. (I), and the experimental points of geneous and heterogeneous pathways of reaction (I).
.
the (J – F)/(J – F) vs. [O ]/[ClO ] plot should be above On the contrary, the iodine signal should markedly
0
3
those predicted by Eq. (5). Figure 5 presents the pro- decrease along the reactor axis if the rate constant for
cessing of the experimental data obtained when C F I termination k is high.
3
7
1c
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

2
08
LARIN et al.
(
J0 – F)/(J – F)
(J – F)/(J – F)
12
0
2
0
8
6
4
2
0
8
6
4
2
0
1
2
3
1
1
0
1
1
1
1
1
1
2
1
4
9
8
7
6
5
4
3
2
1
0
3
1
00
200
300
400
[
500
O ]/[ClO ]
100
200
300
•
[O ]/[ClO ]
3
•
3
Fig. 4. Experimental data (points) obtained with C F I used
3
7
Fig. 5. Results of modeling for the simultaneous occur-
as an iodine atom source at [O ] = 5.0 × 10 molecule/cm3,
14
3
rence of homogeneous reaction (I) and chain termination
P = 2.54 Torr, v = 488 cm/s, and T = 303 K, as well as the
–11
3
–1 –1
(
Ic) at k = 5.0 × 10
cm molecule
1011 cm molecule s–1: (1) 2, (2) 1.5, (3) 1, and (4) 0.
The points represent experimental data for C F I as an
s
and k1c ×
1
results of mathematical modeling including the heteroge-
3
–1
het
neous process (I ) at the degree of reaction homogeneity
het
1
3 7
(
k /(k + k )) equal to (1) 0, (2) 0.2, and (3) 1. The dotted
1 1
iodine atom source; the dotted line corresponds to a 99%
confidence interval.
line corresponds to a 95% confidence interval.
In the case of reaction (Ic), the dependence of the source. Similar relations were also obtained for other
.
.
ss
resonance fluorescence signal of iodine atoms on the
contact time is described by the equation
[
IO ] /([IO ] +[I] ) ratios.
ss ss
The slope of the curve in the ln(J – F) – t coordinates
is keff [IO ] /([IO ] + [I] ). Figure 7 presents the
ss ss
.
.
ss
k [O ]
2
.
3
ln(J – F) = –k --------------------------------------------t
(6)
eff
.
.
.
.
k [ClO ] + k [O ]
k [IO ] /([IO ] + [I] ) vs. [IO ] /([IO ] + [I] )
eff ss ss ss ss ss
1
2
3
ss
plot. The slope of this curve is k . We should know the
eff
or
k value to determine the rate constant k .
5
1c
.
[
IO ]ss
As noted above, the reactor surface was covered
ln(J – F) = –k ------------------------------ -- t,
(7)
(8)
eff
.
with an F32-L Teflon compound before the experiment
[
IO ]ss + [I]ss
.
.
to reduce IO decay. To be sure that the heterogeneous
where
decay of IO radicals is insignificant, we used their
.
reaction with NO, which was thoroughly studied by
resonance fluorescence of iodine atoms in [17]. We
showed in that work that only the homogeneous reac-
tion
keff = k [ClO ] + k
1b
5
.
and k is the rate constant for IO decay on the reactor
5
.
.
walls. The [IO ] /([IO ] + [I] ) can be calculated
using the J value (the resonance fluorescence signal of
k6
ss
ss
ss
IO + NO
I + NO2
(VI)
0
.
.
the iodine atoms in the absence of ClO + O ) and the occurs on addition of NO to IO radicals produced by
3
resonance fluorescence signal of iodine atoms in the reaction (II).
.
.
presence of ClO and O at t
reaction may be neglected.
0, when termination
3
The only way that IO radicals disappear in the
absence of their homogeneous decay is their decay on
Figure 6 illustrates the ln(J – F) vs. t dependence for the reactor walls. If the k value is very low, as in the
5
the experiment with C F I used as an iodine atom case of the reactor coverage with the F32-L Teflon com-
3
7
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

EXPERIMENTAL EVIDENCE FOR ACCELERATION OF REACTION
209
•
•
ln(J – F)
5
k [IO ] /([IO ] + [I] )
eff
ss
ss
ss
4
0
4
3
30
2
1
0
0
0
0.2
0.4
0.6
0.8
2
0
•
•
0.01
0.02
0.03
0.04
t, s
[IO ]ss/([IO ]ss + [I]ss)
.
.
.
.
Fig. 7. keff [IO ] /([IO ] + [I] ) vs. [IO ] /([IO ]
+
Fig. 6. ln(J – F) vs. time of the experiment with C F I used
ss
ss
ss
ss
ss
3
7
[
I] ).
as an iodine atom source.
ss
pound, the iodine atom concentration does not change 0.2; that is, the rate constant of the homogeneous reac-
along the reactor. Similar experiments were performed tion (I) is
.
.
after studying the reaction of IO and ClO radicals.
The obtained dependence of ln(J – F) on the contact
k1 ≤ 1 × 10–12 cm molecule s .
3
–1 –1
time (t = L/v, where L is the reaction zone length) sug- This value agrees well with the data of [7, 8].
.
gests that the heterogeneous decay of IO radicals is
Thus, our data support the findings reported in [7, 8]
that the homogeneous reaction of iodine monoxide
with chlorine monoxide occurs too slowly to confirm
the calculations of Solomon et al. [6], who assumed
that this reaction is very important in stratospheric
ozone depletion, especially taking into account that the
insignificant:
–
1
k = (7.6 ± 5.2) s .
5
.
Using Eq. (8) and knowing the concentration [ClO ] =
.3 × 10 molecule/cm and the values of k and k , we ClO and IO concentrations are too low to ensure this
.
.
1
2
3
8
eff 5
can determine the rate constant for reaction (Ic):
process at even lower rate constants of this reaction [9].
The possible role of much faster heterogeneous pro-
cesses requires further investigation.
–
12
3
–1 –1
k = (5.9 ± 1.5) × 10 cm molecule s .
1
c
Note that this value is the upper limit of the rate con-
stant of chain termination in our reactor. This is due to
the fact that, although the rate constant for the hetero-
geneous decay of iodine atoms was determined before
and after experiments and its values perfectly agreed, it
could be higher in the course of the experiment because
of the treatment of the reactor walls by the reaction
products. This means that we obtained a lower value of
the rate constant for chain termination.
REFERENCES
1. Pickering, K., Thompson, A.M., Scala, J.R., Tao, W.K.,
and Simpson, J., J. Atmos. Chem., 1992, vol. 14, p. 297.
2
. Danielsen, E.F., J. Geophys. Res., 1993, vol. 98, p. 8665.
3
. Kritz, M.A., Rosner, S.W., Kelly, K.K., Zocwenstein, M.,
and Chan, K.R., J. Geophys. Res., 1993, vol. 97, p. 8725.
4
. Chatfield, R.B. and Crutzen, P.J., J. Geophys. Res., 1990,
vol. 95, no. 18, p. 22319.
Nevertheless, even our rate constant for the termina-
tion reaction k is too low to explain the experimental
1
c
5
. McCormick, 1992, WMO/UNEP.
data presented in Fig. 5. Therefore, reaction (I) in our
reactor occurs mainly on the reactor walls and its con-
tribution is maximal if the reactor wall is treated by the
reaction products. Figure 4 shows that the fraction of
the homogeneous component of reaction (I) is at most
6. Solomon, S., Garcia, R.R., and Ravishankara, A.R.,
J. Geophys. Res., 1994, vol. 99, no. 10, p. 20491.
7. Bejanian, Y., Bras, G.L., and Poulet, G., Phys. Chem.,
1997, vol. 101, p. 4088.
KINETICS AND CATALYSIS Vol. 44 No. 2 2003

2
10
LARIN et al.
8
. Turnipseed, A.A., Gilles, M.K., Burkholder, J.B., and 14. Larin, I.K., Messineva, N.A., Nevozhai, D.V., Spasskii, A.I.,
Ravishankara, A.R., J. Phys. Chem., 1997, vol. 101,
and Trofimova, E.M., Kinet. Katal., 2000, vol. 41, no. 3,
p. 5517.
p. 346.
9
. J. Geophys. Res. A, 1997, vol. 102, no. 7, p. 8887.
15. Buben, S.N., Larin, I.K., Messineva, N.A., and Trofi-
1
1
0. Rowland, F.S., Sato, H., Khwaya, H., and Elliott, S.M.,
mova, E.M., Khim. Fiz., 1990, vol. 9, no. 1, p. 116.
J. Phys. Chem., 1985, vol. 90.
1
6. De More, W.B., Sander, S.P., Golden, D.M., Hampson, R.F.,
Kurylo, M.J., Howard, C.J., Ravishankara, A.R., Kolb, C.E.,
and Molina, M.J., Chemical Kinetics and Photochemical
Data for Use in Stratospheric Modeling. Pasadena, CA,
NASA, JPL, California Institute of Technology, 1994.
1. Molina, M.J., Tai-Ly Tso, Molina, L.T., and Wang, F.C.-Y.,
Science, 1988, vol. 238, p. 1253.
1
2. Rown, S., Ozone Crisis, New York: Wiley, 1990.
1
3. Larin, I.K., Messineva, N.A., Spasskii, A.I., Trofi-
mova, E.M., and Turkin, L.E., Kinet. Katal., 2000, 17. Buben, S.N., Larin, I.K., Messineva, N.A., and Trofi-
vol. 41, no. 4, p. 485.
mova, E.M., Khim. Fiz., 1996, vol. 15, no. 1, p. 116.
KINETICS AND CATALYSIS Vol. 44 No. 2 2003